[rrd-users] RRDTool not accepting any updates (NaN instead)

[Tobias Oetiker]

Today Wouter van Eekelen wrote:

> I'm storing the amount of bits that the interface is counting, this
> increases until the integer wrap (hence the high upper limit).
> What would you suggest to store instead? The amount of bits per second at
> that time?

the counter data is fine ... BUT the upper limit in the DS argument is
about the rate not about the counter ...

internally rrdtool calculates the rate and then it applies the
check for upper and lower limit ... 0 is the smallest rate to be
expected and 100MBit/s or 1GBit/s is the highest rate to be
expected ...

cheers
tobi



>
> On Sun, Dec 5, 2010 at 9:29 AM, Tobias Oetiker <tobi at oetiker.ch> wrote:
>
> > Hoi Wouter,
> >
> > Yesterday Wouter van Eekelen wrote:
> >
> > > And just to confirm, these arguments:
> > > --
> > >                 $command = "/usr/bin/rrdtool create ".$rrd_file." -s 60 \
> > >                                 DS:ds0:DERIVE:90:0:4294967295 \
> > >                                 DS:ds1:DERIVE:90:0:4294967295 \
> > >                                 RRA:AVERAGE:0.5:1:600 \
> > >                                 RRA:AVERAGE:0.5:6:700 \
> > >                                 RRA:AVERAGE:0.5:24:775 \
> > >                                 RRA:AVERAGE:0.5:288:797 \
> > >                                 RRA:MAX:0.5:1:600 \
> > >                                 RRA:MAX:0.5:6:700 \
> > >                                 RRA:MAX:0.5:24:775 \
> > >                                 RRA:MAX:0.5:288:797 \
> > >                                 RRA:MIN:0.5:1:600 \
> > >                                 RRA:MIN:0.5:6:700 \
> > >                                 RRA:MIN:0.5:24:775 \
> > >                                 RRA:MIN:0.5:288:797";
> > > --
> > >
> > > Will be fine for graphing bandwidth (on HP procurve switchports)?
> > > Any int wraps (the HP resets to 0 when it passes the unsigned integer
> > limit
> > > of 4,294,967,295) will be recognized and not cause any problems?
> > > How long will the data be stored for?
> >
> > note that the upper limit is the rate and not the actual value you
> > are putting in ... so the 4 trillions are a bit much ... I would
> > rather suggest to use the actual bandwidth of the switch port
> >
> > cheers
> > tobi
> >
> > >
> > > On Sat, Dec 4, 2010 at 10:28 PM, Wouter van Eekelen <
> > > wouter.van.eekelen at serverffs.com> wrote:
> > >
> > > > That works a lot better indeed. Is there a reason updatev can't output
> > such
> > > > information?
> > > > If it would just tell me 'mrhb expired, assuming nan' I would have
> > solved
> > > > this way faster.
> > > >
> > > >
> > > > On Sat, Dec 4, 2010 at 10:08 PM, Tobias Oetiker <tobi at oetiker.ch>
> > wrote:
> > > >
> > > >> Wouter,
> > > >>
> > > >> Today Wouter van Eekelen wrote:
> > > >>
> > > >> > Aren't my updates incremental? The second update has quite a higher
> > > >> value
> > > >> > than the first one, yet it's still NaN.
> > > >>
> > > >> it seems to me that you have set the mrhb to 5 seconds ... while
> > > >> the step size is 60 seconds ... this is possible but it requires
> > > >> you to update the rrd at least every 5 seconds ...
> > > >>
> > > >> cheers
> > > >> tobi
> > > >>
> > > >> >
> > > >> > On Sat, Dec 4, 2010 at 9:23 PM, Steve Shipway <
> > s.shipway at auckland.ac.nz
> > > >> >wrote:
> > > >> >
> > > >> > >   Your DS are of type 'derive', with a minimum value of 0.
> >  'Derive'
> > > >> > > stores the rate of change of the variable, IE (( x -
> > > >> x(prev))/(t-t(prev)))
> > > >> > > This means you'll need at least two sequential updates, which must
> > be
> > > >> > > increasing (in order to get a valid positive derivitive) before
> > you'll
> > > >> see
> > > >> > > anything stored in the RRAs.  Since your step is 60s, this means
> > two
> > > >> > > increasing updates about 1min apart, and more than 2 (4, 13, 145)
> > if
> > > >> you
> > > >> > > want any of the RRAs with a cdp of more than 1 to show a value
> > (since
> > > >> your
> > > >> > > XFF is 0.5).
> > > >> > >
> > > >> > > Steve
> > > >> > >
> > > >> > >  *Steve Shipway*
> > > >> > > University of Auckland ITS
> > > >> > > *UNIX Systems Design Lead*
> > > >> > > s.shipway at auckland.ac.nz
> > > >> > > Ph: +64 9 373 7599 ext 86487
> > > >> > > **
> > > >> > >   ------------------------------
> > > >> > > *From:* rrd-users-bounces+s.shipway=auckland.ac.nz@
> > lists.oetiker.ch
> > > >> [rrd-users-bounces+s.shipway=
> > > >> > > auckland.ac.nz at lists.oetiker.ch] on behalf of Wouter van Eekelen
> > [
> > > >> > > wouter.van.eekelen at serverffs.com]
> > > >> > > *Sent:* Sunday, 5 December 2010 8:09 a.m.
> > > >> > > *To:* rrd-users at lists.oetiker.ch
> > > >> > > *Subject:* [rrd-users] RRDTool not accepting any updates (NaN
> > instead)
> > > >> > >
> > > >> > >  I'm having a huge trouble with RRDtool, all updates I'm doing are
> > > >> > > accepted as 'NaN'
> > > >> > > Here is are all commands with output:
> > > >> > >
> > > >> > >
> > > >> >
> > > >>
> > > >> --
> > > >> Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland
> > > >> http://it.oetiker.ch tobi at oetiker.ch ++41 62 775 9902 / sb: -9900
> > > >>
> > > >
> > > >
> > >
> >
> > --
> > Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland
> > http://it.oetiker.ch tobi at oetiker.ch ++41 62 775 9902 / sb: -9900
> >
>

-- 
Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland
http://it.oetiker.ch tobi at oetiker.ch ++41 62 775 9902 / sb: -9900